Index of content:
Volume 22, Issue 1, 01 January 1951
22(1951); http://dx.doi.org/10.1063/1.1699815View Description Hide Description
The theory presented in this article was developed in an attempt to describe the observed motion and plastic deformation of clamped metal diaphragms used in certain underwater explosion experiments and in certain mechanical gauges. The theoretical attack on this problem enables one to set up certain equations of motion, which may be solved in finite form under certain conditions. The solutions enable one to specify, for instance, the final deformed diaphragm profile, the distribution of thickness after deformation, the swing‐time, which is the total time for deformation to take place, and many other quantities.
The simplest case, termed the ``elementary approximation,'' turns out, except for relatively minor details, to describe adequately for many purposes the motion and final shape of the diaphragm; it is found that the deformed diaphragm shape is conical, the thickness distribution shows a marked dimpling at the center of the diaphragm, and the swing‐time ts is, to this order of approximation , where a is the radius of the diaphragm, and c is the square root of the ratio of the ``yield stress'' to the density. These results are all in good agreement with experimental facts.
22(1951); http://dx.doi.org/10.1063/1.1699812View Description Hide Description
The article shows how lines of force of various magnetic fields can be visually depicted in three dimensions by means of a gas discharge. The method is illustrated by several photographs proving especially its applicability when investigating fields of small geometrical dimensions.
22(1951); http://dx.doi.org/10.1063/1.1699813View Description Hide Description
A theoretical examination is made of the problem of the electromagnetic fields of electric dipoles in the presence of an infinitely long, perfectly conducting cylinder. Application of the Green's function method to this problem yields expressions for the fields in terms of an integration in the complex plane. Simplified expressions for far zone fields are obtained through the known result of the complex integration for the far zone conditions. Results are given for both circular and elliptic cylinders. In the latter case, it is possible to treat the problem of dipoles in the presence of a ground plane which is finite in one dimension and infinite in the other, since this is equivalent to an elliptic cylinder with a vanishing minor axis. Polar diagrams of calculated far zone fields are included for a number of cases.
22(1951); http://dx.doi.org/10.1063/1.1699814View Description Hide Description
The polynomial formulation of the linear array problem introduced by Schelkunoff is studied from the point of view of potential theory. An electrical analogy results which is not only very helpful in qualitative thinking about the problem, but which also makes possible the construction of an analog computer for the synthesis of arrays to produce a desired radiation pattern, and for the evaluation of existing arrays. Such a computer has been built, and tested on known patterns with remarkable success.
Several theorems concerning the effect of the roots of the polynomial on the magnitude and phase of the space factor of an array are given. Some of the most important consequences of these are the lack of uniqueness in the array synthesis problem, and the evaluation of known methods of synthesis which consider only constant‐phase patterns.
A method of analysis is sketched which in principle does the same thing as the computer, and the difficulties in applying this are described. It is concluded that the computer would be preferable for a wide range of practical problems.
22(1951); http://dx.doi.org/10.1063/1.1699816View Description Hide Description
The circular diffraction antenna consists of a coaxial wave guide fitted with an infinite‐plane conducting baffle, and open to free space. An equivalent circuit description, appropriate to principal‐mode propagation in the coaxial region, is investigated theoretically. Variational expressions for the circuit parameters are derived, and used for accurate numerical evaluation.
22(1951); http://dx.doi.org/10.1063/1.1699817View Description Hide Description
Present digital storage devices use two space coordinates or time and one space coordinate for selection switching, resulting in bulky construction or long access time. Three‐dimensional arrays with efficient high speed selection appear possible after continued development of rectangular‐hysteresis magnetic materials. An operating mode is suggested which depends on ability of the magnetic material to discriminate between two values of magnetizing force which differ by a 2:1 ratio. Only one magnetic core per binary digit is required. Tests show that most existing metallic magnetic materials switch in 20 to 10,000 microseconds and are too slow. Nonmetallic magnetic materials can now approach the required magnetic behavior; they switch in less than a microsecond.
22(1951); http://dx.doi.org/10.1063/1.1699818View Description Hide Description
22(1951); http://dx.doi.org/10.1063/1.1699819View Description Hide Description
This paper outlines a theory of excitation of oscillations under the rather slow variations of a parameter on which a system depends. The basic equation is a Mathieu equation. The basic method consists in reducing the system to polar coordinates and applying perturbations.
22(1951); http://dx.doi.org/10.1063/1.1699820View Description Hide Description
The geometrical optical arguments underlying the equivalent earth's radius approximation are extended to non‐horizontal rays. The appropriate wave equation for a non‐uniform, but spherically symmetric, region is derived in a natural way. The notion of a ``primary field'' with the form eikR/R at the source is dropped in favor of a solution to the wave equation in terms of a Green's Function. The formal solution thus obtained is seen to contain the known solutions for a uniform medium. For the non‐uniform case, the solutions to the radial equation are found by a technique due to Langer. These functions account for the index variation near the earth without making unwarranted assumptions about the behavior at greater heights. The formal series obtained is summed by the Watson technique. The first term in this series (lowest mode) alone determines the field at large distances and indicates that one may account for standard atmospheric refraction by using an effective earth's radius.
22(1951); http://dx.doi.org/10.1063/1.1699821View Description Hide Description
The experimental pressure distribution data from low speed wind tunnel tests at zero angle of attack on two different conical noses attached to a cylindrical after‐body are compared with theory. The simplified theoretical equation used is a higher order approximation valid up to the order r 4 lnr. It is shown that the conical nose with a vertex angle of 20 degrees is in agreement but the larger cone with a vertex angle of 30 degrees shows a discrepancy between the measured pressures and the approximate theory. In both cases, however, the theory is in excellent agreement with the surface pressures on the cylindrical after‐body.
22(1951); http://dx.doi.org/10.1063/1.1699822View Description Hide Description
A graphical determination of the position of the Fermi level in a simple impurity semiconductor at zero space charge is discussed. Measurements by G. Busch and H. Labhard indicate that the effective masses of electrons and holes in SiC crystals are of the order of mass of a free electron.
22(1951); http://dx.doi.org/10.1063/1.1699823View Description Hide Description
This article justifies the fact that the fields in a wave guide can be expanded as a linear combination of normal modes. The proof consists in showing that (a) this expansion is possible for the z components of the fields and modes and (b) that, in three dimensions, the fields as a whole are equal to a linear expansion of modes with precisely the same coefficients as were found in the expansion of the z components.
22(1951); http://dx.doi.org/10.1063/1.1699824View Description Hide Description
Several refractories of the nitride, boride, and carbide families are investigated as sources of thermionic electrons. An activation pattern qualitatively common to several of the compounds is observed. Emission data for the final state of activation are presented. Data concurrently obtained for thoria are also given for comparison.
True temperature used in the Richardson equation is obtained by correction from brightness temperature. Temperature correction curves have been obtained. These curves can be satisfactorily represented in the region of thermionic emission by one value of spectralemissivity, which values are listed for the materials studied.
22(1951); http://dx.doi.org/10.1063/1.1699825View Description Hide Description
Diffusion in solids is known to occur along grain boundaries and over free surfaces more rapidly than through the interiors of crystals. In order to facilitate quantitative investigation of grain boundary and surfacediffusion, a mathematical analysis of the problem has been completed, assuming that grain boundarydiffusion is analogous to the diffusion of heat along a thin copper foil imbedded in cork. The calculated diffusion‐penetration relationship for grain boundarydiffusion is shown to agree with the experimentally determined grain boundary self‐diffusion of silver.
Secondary Emission of Nickel‐Barium Mixtures and Rhenium When Bombarded by Electrons with Energies from 50 to 8000 Electron‐Volts22(1951); http://dx.doi.org/10.1063/1.1699826View Description Hide Description
The experimental tube contains a contamination‐free gun which directs an electron beam at normal incidence onto a small thin sheet target which may be heated by conduction. The target is located at the center of a conducting sphere to permit a measure of the secondary current, and also of the energy distribution of the secondary electrons by means of the retarding potential method.
The maximum value of δ, the secondary emission ratio, for a cast alloy of nickel containing 1.5 percent barium is 2.8 and occurs at 800 to 900 volts after a critical amount of previous heat treatment of target. The value of δmax increases with heat treatment of target, subsequent to baking of tube, to the above value and then decreases with further heating. For the critical condition the value of δ above 4000 volts is sensitive to the amount of previous electron bombardment of the target. A δmax of 1.3 for rhenium after thorough out‐gassing was obtained at 900 volts.
The most probable energy of low speed secondary electrons from rhenium is approximately 5 ev.
22(1951); http://dx.doi.org/10.1063/1.1699827View Description Hide Description
The behavior of oxide cathodescoated on a nickel alloy containing about 0.1 percent silicon has been investigated under 100 percent and zero duty‐factor operation. The operation of these cathodes at zero duty factor was found to be highly favorable to the development of a high resistance in a layer of Ba2SiO4 located at the interface between the oxide and the core. The layer itself was present not only in cathodes that had aged without electron emission and had developed interface resistance, but also in cathodes that had aged with electron emission and had not developed interface resistance. A model is presented to account for the experimental observations. Evidence is also presented which shows that an ``active'' nickel alloy that does not lead to undue interface resistance formation may exist.
22(1951); http://dx.doi.org/10.1063/1.1699829View Description Hide Description
A novel method is described of determining the energy distribution of emitted secondaries from an insulator. This is based on an analysis of the transient resulting from pulse bombardment. The analysis is simplest when leakage through the target is negligible, but the effect of leakage is also treated. Space charge limitation of the emitted current is assumed to be negligible.
22(1951); http://dx.doi.org/10.1063/1.1699830View Description Hide Description
An accurate and sensitive method for measuring the complex dielectric constant (ε′‐iε″) of gases in the microwave region is described and critically investigated for sources of error. A resonant‐cavity method is used in which the cavity response curve is displayed on a cathode‐ray tube. The variation of resonance frequency and Q of the cavity when filled with gas are determined from measurements with two frequency markers. The equipment is operated at 9000 Mc, and cavities with Q's of 1×104 are used. In the determination of ε′‐1 (dimensionless) of very low loss gases, an accuracy of 0.4 percent and a sensitivity of 4×10−7 can be obtained. In the determination of the loss factor ε″, an accuracy of 2 percent and a sensitivity of 5×10−7 can be obtained. This method is useful for measuring dispersion and absorption in solids and liquids as well as gases.
Experimental results are given and briefly discussed for ε′‐1 of O2, N2, CO2, He, and air, and for ε″ of NH3 as a function of pressure up to 20 cm Hg.